AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Towards Unsupervised Domain Generalization for Face Anti-Spoofing

Yuchen Liu, Yabo Chen, Mengran Gou, Chun-Ting Huang, Yaoming Wang, Wenrui Dai, Hongkai Xiong

ICCV 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Augmentation and mixture-of-experts generalization · 55 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Split-Rotation-Merge module (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 17.93 -> 12.18Improved
add in-domain nearest-neighbor positives (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 14.31 -> 12.18Improved
add cross-domain nearest-neighbor positives (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 16.08 -> 12.18Improved
add domain-specific normalization before cross-domain NN search (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 13.94 -> 12.18Improved
add merged local features to cross-domain NN distance (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 12.97 -> 12.18Improved
use SRM instead of patch shuffle augmentation (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 15.67 -> 12.18Improved
increase number of nearest neighbors k from 1 to 4 (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 12.18 -> 14.24Hurt
add unlabeled CelebA-Spoof data (200k images) to pretraining (ablation)UDG-FAS · FAS I&C&M to OHTER (%): 12.18 -> 9.06Improved
use UDG-FAS unsupervised pretraining instead of ImageNet initResNet-18 FAS · UDG-Protocol-1average AUC: +9.43% average AUCImproved
use UDG-FAS instead of transformer SSL (MAE) for pretrainingFAS pretraining · FAS I&C&M to OHTER (%): 15.28 -> 12.18Improved
combine UDG-FAS pretraining with SSDG finetuningSSDG-R · UDG-Protocol-1average HTER: 11.29% -> 8.15%Improved
add UDG-FAS to SimSiam-style contrastive learning vs generic methods (MoCo, SimCLR, BYOL)unsupervised pretraining · UDG-Protocol-1 FASHTER (%): I&C&M to O: 26.67 (SimCLR V2) -> 12.18Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy